Search

ISO 26262 requires classifying random hardware faults based on their effects (safe, detected, or undetected) within integrated circuits used in automobiles. In general, this classification is addressed using expert judgment and a combination of tools. However, the growth of integrated circuit complexity creates a huge fault space; hence, this form of fault classification is error prone and time consuming. Therefore, an automated and systematic approach is needed to target hardware fault classification in automotive systems on chips (SoCs), considering the application software. This work focuses on identifying safe faults: the proposed approach utilizes coverage analysis to identify candidate safe faults considering all the constraints coming from the application. Then, the behavior of the application software is modeled so that we can resort to a formal analysis tool. The proposed technique is evaluated on the AutoSoC benchmark running a cruise control application. Resorting to our approach, we could classify 20%, 11%, and 13% of all faults in the central processing unit (CPU), universal asynchronous receiver–transmitter (UART), and controller area network (CAN) as safe faults, respectively. We also show that this classification can increase the diagnostic coverage of software test libraries targeting the CPU and CAN modules by 4% to 6%, increasing the achieved testable fault coverage., Computer Engineering, Quantum & Computer Engineering

Fault detection, identification(classification), and isolation are very important to ensure continuous power transmission in a power grid system. Fast and accurate methods are thus very critical for fault diagnosis in power systems. Many research papers attempted different learning methods in this area, and the focus was either on a subset of the aspects or only on the critical asset faults. The key results from some of the relevant ones are taken up for comparison. This paper describes the novel technical results in detecting and identifying all types of AC and DC faults in the HVDC station by using a fully convolutional neural network (FCNN) deep learning algorithm. The performance is evaluated with an experiment on symmetrical monopolar HVDC station simulated in Power Systems Computer-Aided Design (PSCAD). The novel significance of the results includes applying the learned knowledge from one station to validate on the other station data, the quick time to detect and identify faults, the confusion matrix, classification reports with probability of 99.24% for detection and 97.73% for identification, False alarm rate of 1.35%, and zero percent missed faults. The adaptability of the trained model from the learned knowledge to schematically related HVDC stations is discussed.

ISO 26262 requires classifying random hardware faults based on their effects (safe, detected, or undetected) within integrated circuits used in automobiles. In general, this classification is addressed using expert judgment and a combination of tools. However, the growth of integrated circuit complexity creates a huge fault space; hence, this form of fault classification is error prone and time consuming. Therefore, an automated and systematic approach is needed to target hardware fault classification in automotive systems on chips (SoCs), considering the application software. This work focuses on identifying safe faults: the proposed approach utilizes coverage analysis to identify candidate safe faults considering all the constraints coming from the application. Then, the behavior of the application software is modeled so that we can resort to a formal analysis tool. The proposed technique is evaluated on the AutoSoC benchmark running a cruise control application. Resorting to our approach, we could classify 20%, 11%, and 13% of all faults in the central processing unit (CPU), universal asynchronous receiver–transmitter (UART), and controller area network (CAN) as safe faults, respectively. We also show that this classification can increase the diagnostic coverage of software test libraries targeting the CPU and CAN modules by 4% to 6%, increasing the achieved testable fault coverage., Computer Engineering, Quantum & Computer Engineering

As the automotive industry constantly makes technological progress, higher demands are placed on safety, environmentally friendly and durability. Modern vehicles are headed towards increasingly complex system, in terms of both hardware and software making it important to detect faults in any of the components. Monitoring the engine’s health has traditionally been done using expert knowledge and model-based techniques, where derived models of the system’s nominal state are used to detect any deviations. However, due to increased complexity of the system this approach faces limitations regarding time and knowledge to describe the engine’s states. An alternative approach is therefore data-driven methods which instead are based on historical data measured from different operating points that are used to draw conclusion about engine’s present state. In this thesis a proposed diagnostic framework is presented, consisting of a systematically approach for fault classification of known and unknown faults along with a fault size estimation. The basis for this lies in using principal component analysis to find the fault vector for each fault class and decouple one fault at the time, thus creating different subspaces. Importantly, this work investigates the efficiency of taking multiple classifiers into account in the decision making from a performance perspective. Aggregating multiple classifiers is done solving a quadratic optimization problem. To evaluate the performance, a comparison with a random forest classifier has been made. Evaluation with challenging test data show promising results where the algorithm relates well to the performance of random forest classifier.

The occurrence of faults in any operational power system network is inevitable, and many of the causative factors such as lightning, thunderstorm among others is usually beyond human control. Consequently, there is the need to set up models capable of prompt identification and classification of these faults for immediate action. This paper, explored the use of artificial neural network (ANN) technique to identify and classify various faults on the 11 kV distribution network of University of Lagos. The ANN is applied because it offers high speed, higher efficiency and requires less human intervention. Datasets of the case study obtained were sectioned proportionately for training, testing, and validation. The mathematical formulations for the method are presented with python used as the programming tools for the analysis. The results obtained from this study, for both the voltage and current under different scenarios of faults, are displayed in graphical forms and discussed. The results showed the effectiveness of the ANN in fault identification and classification in a distribution network as the model yielded satisfactory results for the available limited datasets used. The information obtained from this study could be helpful to the system operators in faults identification and classification for making informed decisions regarding power system design and reliability.

The occurrence of faults in any operational power system network is inevitable, and many of the causative factors such as lightning, thunderstorm among others is usually beyond human control. Consequently, there is the need to set up models capable of prompt identification and classification of these faults for immediate action. This paper, explored the use of artificial neural network (ANN) technique to identify and classify various faults on the 11 kV distribution network of University of Lagos. The ANN is applied because it offers high speed, higher efficiency and requires less human intervention. Datasets of the case study obtained were sectioned proportionately for training, testing, and validation. The mathematical formulations for the method are presented with python used as the programming tools for the analysis. The results obtained from this study, for both the voltage and current under different scenarios of faults, are displayed in graphical forms and discussed. The results showed the effectiveness of the ANN in fault identification and classification in a distribution network as the model yielded satisfactory results for the available limited datasets used. The information obtained from this study could be helpful to the system operators in faults identification and classification for making informed decisions regarding power system design and reliability.

As the automotive industry constantly makes technological progress, higher demands are placed on safety, environmentally friendly and durability. Modern vehicles are headed towards increasingly complex system, in terms of both hardware and software making it important to detect faults in any of the components. Monitoring the engine’s health has traditionally been done using expert knowledge and model-based techniques, where derived models of the system’s nominal state are used to detect any deviations. However, due to increased complexity of the system this approach faces limitations regarding time and knowledge to describe the engine’s states. An alternative approach is therefore data-driven methods which instead are based on historical data measured from different operating points that are used to draw conclusion about engine’s present state. In this thesis a proposed diagnostic framework is presented, consisting of a systematically approach for fault classification of known and unknown faults along with a fault size estimation. The basis for this lies in using principal component analysis to find the fault vector for each fault class and decouple one fault at the time, thus creating different subspaces. Importantly, this work investigates the efficiency of taking multiple classifiers into account in the decision making from a performance perspective. Aggregating multiple classifiers is done solving a quadratic optimization problem. To evaluate the performance, a comparison with a random forest classifier has been made. Evaluation with challenging test data show promising results where the algorithm relates well to the performance of random forest classifier.

Inhabited isolated areas depend on independent power systems like islanded microgrids for their electricity supply. These islanded microgrids require adaptive and rapid disturbance detection techniques due to their insufficient kinetic energy reserve and quick response of power electronic inverters of distributed renewable energy sources. The overreaching goals of this thesis is to explore such issues in islanded AC microgrids and develop near real-time disturbance detection and protective solutions for their stable operation. As the first contribution of this thesis, a centralized, voltage signal-based fault detection and classification method is proposed using digital signal processing tools and machine learning techniques for islanded AC microgrids. The features used for fault detection and classification are instantaneous jumps in voltage amplitude, phase angle and frequency. This method has the capability to detect any kind of system disturbance and can differentiate the faults from non-fault disturbances in both balanced and unbalanced systems. Fault detection performance accuracies are over 99%, 82% and 72% for the faults up to 10Ω, 50Ω and 100Ω respectively, and fault classification performance accuracy is over 95% for the faults up to 100Ω.Secondly, an adaptive, decentralized, differential feature-based protection scheme is proposed for islanded AC microgrids. Instead of differential currents, differential features are used, and edge computing is utilized to extract the features at the point of measurement. The most informative one-dimensional line differential feature is compared with an adaptive threshold to perform the fault detection. The threshold is updated every second and is immune to system noises. Fault classification is performed with a simple logical comparison. An in-memory circular buffer is proposed to store the local features in case of communication delay. Communication and synchronization are experimentally validated. This scheme produces

This paper studied the use of a new stator current feature for detection of winding and cage bars faults in an induction motor, and presents the experimental validation of a detection and identification scheme using Support Vector Machines (SVM). This validation was performed in a test bed using 2 HP, 4 pole motors in which shorted winding and broken bars faults were induced, separately. Both time and frequency domain features like arithmetic mean, RMS value, Central Frequency, Kurtosis, RMS value of Power Spectral Density were assessed and validated using experimental data for several load conditions. PSC/NSC (positive sequence current/ negative sequence current) ratio was successful in ' most of the classifiers despite the load regime. This new feature was evaluated in terms of fault detection and severity discrimination with satisfactory results., Resumo Neste trabalho foi estudado o uso de um novo recurso para a detecção de falhas nos rolamentos e na gaiola de esquilo de um motor de indução, e apresenta a validação experimental de um esquema de detecção e identificação utilizando máquinas de vetores de suporte (SVM). Esta validação foi realizada num banco de ensaios utilizando motores de 2 HP, 4 polos nos quais foram induzidas as falhas de curto circuito no estator e barras quebradas, separadamente. Os recursos em domínio do tempo e no domínio da frequência como a média aritmética, valor RMS, frequência central, curtose, valor RMS da densidade espectral de potência foram avaliados e validados a partir de dados experimentais para várias condições de carga. O recurso relação PSC/NSC (corrente de sequência positiva/corrente de sequência negativa) teve um desempenho satisfatório na maioria dos classificadores independente do régime de carga. Este novo recurso foi avaliada em termos de detecção de falhas e discriminação entre os diferentes graus de severidade com resultados satisfatórios., Resumen En este trabajo se estudió el uso de un nuevo rasgo para la detección de fallas en el devanado y la jaula de un motor de inducción, y presenta la validación experimental de un esquema de detección e identificación utilizando Máquinas de Vectores Soporte (SVM). Esta validación se realizó en un banco de pruebas usando motores de 2 HP, 4 polos en los que fueron inducidas las fallas de corto circuito en el estator y barras rotas, por separado. Los rasgos en dominio tiempo y dominio de la frecuencia como la media aritmética, valor RMS, frecuencia central, curtosis, valor RMS de la densidad espectral de potencia fueron evaluados y validados a partir de datos experimentales para varias condiciones de carga. El rasgo relación PSC/NSC (corriente de secuencia positiva / corriente de secuencia negativa) tuvo un desempeño satisfactorio en la mayoría de los clasificadores independiente del régimen de carga. Este nuevo rasgo se evaluó en términos de detección de fallos y la discriminación entre los diferentes grados de severidad con resultados satisfactorios.

This paper studied the use of a new stator current feature for detection of winding and cage bars faults in an induction motor, and presents the experimental validation of a detection and identification scheme using Support Vector Machines (SVM). This validation was performed in a test bed using 2 HP, 4 pole motors in which shorted winding and broken bars faults were induced, separately. Both time and frequency domain features like arithmetic mean, RMS value, Central Frequency, Kurtosis, RMS value of Power Spectral Density were assessed and validated using experimental data for several load conditions. PSC/NSC (positive sequence current/ negative sequence current) ratio was successful in ' most of the classifiers despite the load regime. This new feature was evaluated in terms of fault detection and severity discrimination with satisfactory results., Resumo Neste trabalho foi estudado o uso de um novo recurso para a detecção de falhas nos rolamentos e na gaiola de esquilo de um motor de indução, e apresenta a validação experimental de um esquema de detecção e identificação utilizando máquinas de vetores de suporte (SVM). Esta validação foi realizada num banco de ensaios utilizando motores de 2 HP, 4 polos nos quais foram induzidas as falhas de curto circuito no estator e barras quebradas, separadamente. Os recursos em domínio do tempo e no domínio da frequência como a média aritmética, valor RMS, frequência central, curtose, valor RMS da densidade espectral de potência foram avaliados e validados a partir de dados experimentais para várias condições de carga. O recurso relação PSC/NSC (corrente de sequência positiva/corrente de sequência negativa) teve um desempenho satisfatório na maioria dos classificadores independente do régime de carga. Este novo recurso foi avaliada em termos de detecção de falhas e discriminação entre os diferentes graus de severidade com resultados satisfatórios., Resumen En este trabajo se estudió el uso de un nuevo rasgo para la detección de fallas en el devanado y la jaula de un motor de inducción, y presenta la validación experimental de un esquema de detección e identificación utilizando Máquinas de Vectores Soporte (SVM). Esta validación se realizó en un banco de pruebas usando motores de 2 HP, 4 polos en los que fueron inducidas las fallas de corto circuito en el estator y barras rotas, por separado. Los rasgos en dominio tiempo y dominio de la frecuencia como la media aritmética, valor RMS, frecuencia central, curtosis, valor RMS de la densidad espectral de potencia fueron evaluados y validados a partir de datos experimentales para varias condiciones de carga. El rasgo relación PSC/NSC (corriente de secuencia positiva / corriente de secuencia negativa) tuvo un desempeño satisfactorio en la mayoría de los clasificadores independiente del régimen de carga. Este nuevo rasgo se evaluó en términos de detección de fallos y la discriminación entre los diferentes grados de severidad con resultados satisfactorios.

The reliability requirements of wind turbines have increased significantly in recent years inthe search for a lower impact on the cost of energy. In addition, the trend towards larger wind turbinesinstalled in remote locations has significantly increased the cost of repair or replacement of the compo-nent. In the wind industry, therefore, condition monitoring is crucial for maximum availability [1]. Thiscontribution makes a review of supervised machine learning classification techniques for wind turbinecondition monitoring using only SCADA data already available. That is, without installing extra sensorsor costly purpose-built data sensing equipment. Although there has been extensive research into the useof machine learning techniques for wind turbine monitoring, the more recent trend in this type of litera-ture is to focus on a specific WT sub-assembly: the bearings and planetary gearbox [2], the generator andpower converter [3], the blades [4], etc. Oil debris systems can detect pitting failures but cannot detectcracking faults. Vibration based systems can detect both pitting and cracking, but most cannot determinethe health of components in the planetary section. This work approaches condition monitoring of variouswind turbine components (torque actuator, pitch actuator, pitch sensor, and generator speed sensor) witha unique strategy. In particular, for this purpose, a review of supervised machine learning classificationtechniques is performed and analyzed., Postprint (published version)

This study is focused on the current challenges dealing with electromechanical system monitoring applied in industrial frameworks, that is, the presence of unknown events and the limitation to the nominal healthy condition as starting knowledge. Thus, an industrial machinery condition monitoring methodology based on novelty detection and classification is proposed in this study. The methodology is divided in three main stages. First, a dedicated feature calculation and reduction over each available physical magnitude. Second, an ensemble structure of novelty detection models based on one-class support vector machines to identify not previously considered events. Third, a diagnosis model supported by a feature fusion scheme in order to reach high fault classification capabilities. The effectiveness of the fault detection and identification methodology has been compared with classical single model approach, and verified by experimental results obtained from an electromechanical machine. © 2018 IEEE., Postprint (author's final draft)

Due to the increasing installation of wind turbines in remote locations, both onshore and offshore, advanced fault detection and classification strategies have become crucial to accomplish the required levels of reliability and availability. In this work, without using specific tailored devices for condition monitoring but only increasing the sampling frequency in the already available (in all commercial wind turbines) sensors of the Supervisory Control and Data Acquisition (SCADA) system, a data-driven multi-fault detection and classification strategy is developed. An advanced wind turbine benchmark is used. The wind turbine we consider is subject to different types of faults on actuators and sensors. The main challenges of the wind turbine fault detection lie in their non-linearity, unknown disturbances, and significant measurement noise at each sensor. First, the SCADA measurements are pre-processed by group scaling and feature transformation (from the original high-dimensional feature space to a new space with reduced dimensionality) based on multiway principal component analysis through sample-wise unfolding. Then, 10-fold cross-validation support vector machines-based classification is applied. In this work, support vector machines were used as a first choice for fault detection as they have proven their robustness for some particular faults, but at the same time have never accomplished the detection and classification of all the proposed faults considered in this work. To this end, the choice of the features as well as the selection of data are of primary importance. Simulation results showed that all studied faults were detected and classified with an overall accuracy of 98.2%. Finally, it is noteworthy that the prediction speed allows this strategy to be deployed for online (real-time) condition monitoring in wind turbines., Postprint (published version)

With the rapid development of power systems, locating and classifying faults is critical to the continuity and reliability of the transmission system. In this thesis, a traveling-wave based technique for fault location and classification on high voltage and extremely high-voltage transmission lines is proposed. The traveling-wave based protection has the advantage of fast response and not being affected by power swing and CTs saturation. In this thesis, the transient characteristics of single line to ground fault (which can be divided into solid fault and arcing fault) and lightning disturbance are extracted by using Clarke transformation and wavelet transformation. The differences among recorded traveling wave arrival times are used to calculate the fault location, and the wavelet energy at different frequency bands is utilized to distinguish between lightning and different kinds of fault. A criterion is proposed according to the energy ratio. The proposed scheme can identify different faults correctly and quickly. In addition, the influence of busbar capacitance, current transformer and coupling capacitor voltage transformer are considered. The simulation of a transmission system has been made in ATP/EMTP, and the calculations have been made in MATLAB., Med den snabba utvecklingen av kraftsystem är lokalisering och klassificering av fel avgörande för kontinuiteten och tillförlitligheten hos överföringssystem. I denna avhandling föreslås en vågrörelse-baserad teknik för fellokalisering och klassificering av kraftledningar för högspänning och extremt hög spänning. Vågrörelsebaserat skydd har fördelen av snabb respons och att det inte påverkas av kraft fluktuationer och strömtransformsmättnad. I denna avhandling tas momentana egenskaperna av jord till ledningsfel (vilket kan delas in i stumt jordfel och ljusbågefel) och blixtstörning fram med hjälp av Clarke transformation och wavelet transformation. Skillnaderna mellan de uppmätta vågrörelsernas ankomsttider används för att beräkna fellokalisering och wavelet energin vid olika frekvensband, vilket används för att skilja mellan blixt och olika sorters fel. Ett kriterium föreslås enligt energiförhållandet. Det föreslagna systemet kan identifiera olika sorters fel korrekt och snabbt. Dessutom övervägs påverkan av strömskenans kapacitans, strömtransformator och kopplingskondensatorspänningsomvandlare. Simuleringen av transmissionssystem har gjorts med ATP/EMTP, och beräkningarna är gjorda med MATLAB.

Network-on-Chip (NoC) is becoming a competitive solution to connect hundreds of processing elements in modern computing platforms. Under the trend of shrinking feature sizes, circuits are likely to suffer from faults which lead to degraded performance and erroneous behaviour. Compared to permanent faults, transient faults happen even more frequently and seriously while they are hidden within complex on chip behaviours. One of the serious consequences caused by transient faults is taking illegal turns by the packets after the damage of control logic in on-chip routers which may lead to a deadlock situation and eventually crashing the entire system. To avoid this situation, in this paper, we propose a comprehensive scheme called ODT including an improved router architecture, an illegal-turn-resilient routing algorithm, online fault-detect units and a fault classification method. By applying ODT, more turns are supported on routing level and the deadlock situations can be significantly reduced. Experimental results indicate up to 22% increase of the survived packets in the network when 4% of routing computation units in failure. The extra area overhead and power consumption of ODT method is around 9.22% and 9.63%., QC 20160711

With the rapid development of power systems, locating and classifying faults is critical to the continuity and reliability of the transmission system. In this thesis, a traveling-wave based technique for fault location and classification on high voltage and extremely high-voltage transmission lines is proposed. The traveling-wave based protection has the advantage of fast response and not being affected by power swing and CTs saturation. In this thesis, the transient characteristics of single line to ground fault (which can be divided into solid fault and arcing fault) and lightning disturbance are extracted by using Clarke transformation and wavelet transformation. The differences among recorded traveling wave arrival times are used to calculate the fault location, and the wavelet energy at different frequency bands is utilized to distinguish between lightning and different kinds of fault. A criterion is proposed according to the energy ratio. The proposed scheme can identify different faults correctly and quickly. In addition, the influence of busbar capacitance, current transformer and coupling capacitor voltage transformer are considered. The simulation of a transmission system has been made in ATP/EMTP, and the calculations have been made in MATLAB., Med den snabba utvecklingen av kraftsystem är lokalisering och klassificering av fel avgörande för kontinuiteten och tillförlitligheten hos överföringssystem. I denna avhandling föreslås en vågrörelse-baserad teknik för fellokalisering och klassificering av kraftledningar för högspänning och extremt hög spänning. Vågrörelsebaserat skydd har fördelen av snabb respons och att det inte påverkas av kraft fluktuationer och strömtransformsmättnad. I denna avhandling tas momentana egenskaperna av jord till ledningsfel (vilket kan delas in i stumt jordfel och ljusbågefel) och blixtstörning fram med hjälp av Clarke transformation och wavelet transformation. Skillnaderna mellan de uppmätta vågrörelsernas ankomsttider används för att beräkna fellokalisering och wavelet energin vid olika frekvensband, vilket används för att skilja mellan blixt och olika sorters fel. Ett kriterium föreslås enligt energiförhållandet. Det föreslagna systemet kan identifiera olika sorters fel korrekt och snabbt. Dessutom övervägs påverkan av strömskenans kapacitans, strömtransformator och kopplingskondensatorspänningsomvandlare. Simuleringen av transmissionssystem har gjorts med ATP/EMTP, och beräkningarna är gjorda med MATLAB.

Modernized and advanced electricity transmission and distribution infrastructure ensures reliable, efficient, and affordable delivery of electric power. The complexity of fault location problem increases with the proliferation of unusual topologies and with the advent of renewable energy-based power generation in the smart grid environment. The proliferation of new Intelligent Electronic Devices (IEDs) provides a venue for the implementation of more accurate and intelligent fault location methods. This dissertation focuses on intelligent fault location methods for smart power grids and it aims at improving fault location accuracies and decreasing the cost and the mean time to repair damaged equipment in major power outages subsequently increasing the reliability of the grid. The developed methods utilize wavelet transformation to extract the traveling wave information in the very fast voltage and current transients which are initiated immediately after a fault occurs, support vector machines to classify the fault type and identify the faulted branches and finally Bewley diagrams to precisely locate the fault. The approach utilizes discrete wavelet transformation (DWT) for analysis of transient voltage and current measurements. The transient wavelet energies are calculated and utilized as the input for support vector machine (SVM) classifiers. SVM learns the mapping between inputs (i.e. transient voltages and/or currents wavelet energies) and desired outputs (i.e. faulty phase and/or faulty section) through processing a set of training cases. This dissertation presents the proposed methodologies applied to three complex power transmission systems. The first transmission system is a three-terminal (teed) three-phase AC transmission network, a common topology in high- and extra high-voltage networks. It is used to connect three substations that are wide apart from each other through long transmission lines with a tee-point, which is not supported by a substation nor eq

This project, conducted at Ericsson AB, investigates the feasibility of implementing machine learning techniques in order to classify dump files for more effi cient trouble report routing. The project focuses on supervised machine learning methods and in particular Bayesian statistics. It shows that a program utilizing Bayesian methods can achieve well above random prediction accuracy. It is therefore concluded that machine learning methods may indeed become a viable alternative to human classification of trouble reports in the near future., Detta examensarbete, utfört på Ericsson AB, ämnar att undersöka huruvida maskininlärningstekniker kan användas för att klassificera dumpfiler för mer effektiv problemidentifiering. Projektet fokuserar på övervakad inlärning och då speciellt Bayesiansk klassificering. Arbetet visar att ett program som utnyttjar Bayesiansk klassificering kan uppnå en noggrannhet väl över slumpen. Arbetet indikerar att maskininlärningstekniker mycket väl kan komma att bli användbara alternativ till mänsklig klassificering av dumpfiler i en nära framtid.

A fault classification method is proposed which has been applied to an electric vehicle. Potential faults in the different subsystems that can affect the vehicle directional stability were collected in a failure mode and effect analysis. Similar driveline faults were grouped together if they resembled each other with respect to their influence on the vehicle dynamic behaviour. The faults were physically modelled in a simulation environment before they were induced in a detailed vehicle model under normal driving conditions. A special focus was placed on faults in the driveline of electric vehicles employing in-wheel motors of the permanent magnet type. Several failures caused by mechanical and other faults were analysed as well. The fault classification method consists of a controllability ranking developed according to the functional safety standard ISO 26262. The controllability of a fault was determined with three parameters covering the influence of the longitudinal, lateral and yaw motion of the vehicle. The simulation results were analysed and the faults were classified according to their controllability using the proposed method. It was shown that the controllability decreased specifically with increasing lateral acceleration and increasing speed. The results for the electric driveline faults show that this trend cannot be generalised for all the faults, as the controllability deteriorated for some faults during manoeuvres with low lateral acceleration and low speed. The proposed method is generic and can be applied to various other types of road vehicles and faults., QC 20140509. Updated from manuscript to article in journalFunder: Swedish Hybrid vehicle Center

This project, conducted at Ericsson AB, investigates the feasibility of implementing machine learning techniques in order to classify dump files for more effi cient trouble report routing. The project focuses on supervised machine learning methods and in particular Bayesian statistics. It shows that a program utilizing Bayesian methods can achieve well above random prediction accuracy. It is therefore concluded that machine learning methods may indeed become a viable alternative to human classification of trouble reports in the near future., Detta examensarbete, utfört på Ericsson AB, ämnar att undersöka huruvida maskininlärningstekniker kan användas för att klassificera dumpfiler för mer effektiv problemidentifiering. Projektet fokuserar på övervakad inlärning och då speciellt Bayesiansk klassificering. Arbetet visar att ett program som utnyttjar Bayesiansk klassificering kan uppnå en noggrannhet väl över slumpen. Arbetet indikerar att maskininlärningstekniker mycket väl kan komma att bli användbara alternativ till mänsklig klassificering av dumpfiler i en nära framtid.

A fault classification method is proposed which has been applied to an electric vehicle. Potential faults in the different subsystems that can affect the vehicle directional stability were collected in a failure mode and effect analysis. Similar driveline faults were grouped together if they resembled each other with respect to their influence on the vehicle dynamic behaviour. The faults were physically modelled in a simulation environment before they were induced in a detailed vehicle model under normal driving conditions. A special focus was placed on faults in the driveline of electric vehicles employing in-wheel motors of the permanent magnet type. Several failures caused by mechanical and other faults were analysed as well. The fault classification method consists of a controllability ranking developed according to the functional safety standard ISO 26262. The controllability of a fault was determined with three parameters covering the influence of the longitudinal, lateral and yaw motion of the vehicle. The simulation results were analysed and the faults were classified according to their controllability using the proposed method. It was shown that the controllability decreased specifically with increasing lateral acceleration and increasing speed. The results for the electric driveline faults show that this trend cannot be generalised for all the faults, as the controllability deteriorated for some faults during manoeuvres with low lateral acceleration and low speed. The proposed method is generic and can be applied to various other types of road vehicles and faults., QC 20140509. Updated from manuscript to article in journalFunder: Swedish Hybrid vehicle Center

A fault classification method is proposed which has been applied to an electric vehicle. Potential faults in the different subsystems that can affect the vehicle directional stability were collected in a failure mode and effect analysis. Similar driveline faults were grouped together if they resembled each other with respect to their influence on the vehicle dynamic behaviour. The faults were physically modelled in a simulation environment before they were induced in a detailed vehicle model under normal driving conditions. A special focus was placed on faults in the driveline of electric vehicles employing in-wheel motors of the permanent magnet type. Several failures caused by mechanical and other faults were analysed as well. The fault classification method consists of a controllability ranking developed according to the functional safety standard ISO 26262. The controllability of a fault was determined with three parameters covering the influence of the longitudinal, lateral and yaw motion of the vehicle. The simulation results were analysed and the faults were classified according to their controllability using the proposed method. It was shown that the controllability decreased specifically with increasing lateral acceleration and increasing speed. The results for the electric driveline faults show that this trend cannot be generalised for all the faults, as the controllability deteriorated for some faults during manoeuvres with low lateral acceleration and low speed. The proposed method is generic and can be applied to various other types of road vehicles and faults., QC 20140509. Updated from manuscript to article in journalFunder: Swedish Hybrid vehicle Center

A fault classification method is proposed which has been applied to an electric vehicle. Potential faults in the different subsystems that can affect the vehicle directional stability were collected in a failure mode and effect analysis. Similar driveline faults were grouped together if they resembled each other with respect to their influence on the vehicle dynamic behaviour. The faults were physically modelled in a simulation environment before they were induced in a detailed vehicle model under normal driving conditions. A special focus was placed on faults in the driveline of electric vehicles employing in-wheel motors of the permanent magnet type. Several failures caused by mechanical and other faults were analysed as well. The fault classification method consists of a controllability ranking developed according to the functional safety standard ISO 26262. The controllability of a fault was determined with three parameters covering the influence of the longitudinal, lateral and yaw motion of the vehicle. The simulation results were analysed and the faults were classified according to their controllability using the proposed method. It was shown that the controllability decreased specifically with increasing lateral acceleration and increasing speed. The results for the electric driveline faults show that this trend cannot be generalised for all the faults, as the controllability deteriorated for some faults during manoeuvres with low lateral acceleration and low speed. The proposed method is generic and can be applied to various other types of road vehicles and faults., QC 20140509. Updated from manuscript to article in journalFunder: Swedish Hybrid vehicle Center

A fault classification method is proposed which has been applied to an electric vehicle. Potential faults in the different subsystems that can affect the vehicle directional stability were collected in a failure mode and effect analysis. Similar driveline faults were grouped together if they resembled each other with respect to their influence on the vehicle dynamic behaviour. The faults were physically modelled in a simulation environment before they were induced in a detailed vehicle model under normal driving conditions. A special focus was placed on faults in the driveline of electric vehicles employing in-wheel motors of the permanent magnet type. Several failures caused by mechanical and other faults were analysed as well. The fault classification method consists of a controllability ranking developed according to the functional safety standard ISO 26262. The controllability of a fault was determined with three parameters covering the influence of the longitudinal, lateral and yaw motion of the vehicle. The simulation results were analysed and the faults were classified according to their controllability using the proposed method. It was shown that the controllability decreased specifically with increasing lateral acceleration and increasing speed. The results for the electric driveline faults show that this trend cannot be generalised for all the faults, as the controllability deteriorated for some faults during manoeuvres with low lateral acceleration and low speed. The proposed method is generic and can be applied to various other types of road vehicles and faults., QC 20140509. Updated from manuscript to article in journalFunder: Swedish Hybrid vehicle Center

The increase of electronics in road vehicles comes along with a broad variety of possibilitiesin terms of safety, handling and comfort for the users. A rising complexityof the vehicle subsystems and components accompanies this development and has tobe managed by increased electronic control. More potential elements, such as sensors,actuators or software codes, can cause a failure independently or by mutually influencingeach other. There is a need of a structured approach to sort the faults from avehicle dynamics stability perspective.This thesis tries to solve this issue by suggesting a fault classification method and faulttolerantcontrol strategies. Focus is on typical faults of the electric driveline and thecontrol system, however mechanical and hydraulic faults are also considered. Duringthe work, a broad failure mode and effect analysis has been performed and the faultshave been modeled and grouped based on the effect on the vehicle dynamic behaviour.A method is proposed and evaluated, where faults are categorized into different levelsof controllability, i. e. levels on how easy or difficult it is to control a fault for the driver,but also for a control system.Further, fault-tolerant control strategies are suggested that can handle a fault with acritical controllability level. Two strategies are proposed and evaluated based on thecontrol allocation method and an electric vehicle with typical faults. It is shown thatthe control allocation approaches give less critical trajectory deviation compared to noactive control and a regular Electronic Stability Control algorithm.To conclude, this thesis work contributes with a methodology to analyse and developfault-tolerant solutions for electric vehicles with improved traffic safety., QC 20131010

This paper discusses how to use the warping information obtained after batch synchronization for process monitoring and fault classification. The warping information can be used for i) building unsupervised control charts or ii) fault classification when a rich faulty batches database is available. Data from realistic simulations of a fermentation process of the Saccharomyces cerevisiae cultivation are used to illustrate the proposal.

The increase of electronics in road vehicles comes along with a broad variety of possibilitiesin terms of safety, handling and comfort for the users. A rising complexityof the vehicle subsystems and components accompanies this development and has tobe managed by increased electronic control. More potential elements, such as sensors,actuators or software codes, can cause a failure independently or by mutually influencingeach other. There is a need of a structured approach to sort the faults from avehicle dynamics stability perspective.This thesis tries to solve this issue by suggesting a fault classification method and faulttolerantcontrol strategies. Focus is on typical faults of the electric driveline and thecontrol system, however mechanical and hydraulic faults are also considered. Duringthe work, a broad failure mode and effect analysis has been performed and the faultshave been modeled and grouped based on the effect on the vehicle dynamic behaviour.A method is proposed and evaluated, where faults are categorized into different levelsof controllability, i. e. levels on how easy or difficult it is to control a fault for the driver,but also for a control system.Further, fault-tolerant control strategies are suggested that can handle a fault with acritical controllability level. Two strategies are proposed and evaluated based on thecontrol allocation method and an electric vehicle with typical faults. It is shown thatthe control allocation approaches give less critical trajectory deviation compared to noactive control and a regular Electronic Stability Control algorithm.To conclude, this thesis work contributes with a methodology to analyse and developfault-tolerant solutions for electric vehicles with improved traffic safety., QC 20131010

The increase of electronics in road vehicles comes along with a broad variety of possibilitiesin terms of safety, handling and comfort for the users. A rising complexityof the vehicle subsystems and components accompanies this development and has tobe managed by increased electronic control. More potential elements, such as sensors,actuators or software codes, can cause a failure independently or by mutually influencingeach other. There is a need of a structured approach to sort the faults from avehicle dynamics stability perspective.This thesis tries to solve this issue by suggesting a fault classification method and faulttolerantcontrol strategies. Focus is on typical faults of the electric driveline and thecontrol system, however mechanical and hydraulic faults are also considered. Duringthe work, a broad failure mode and effect analysis has been performed and the faultshave been modeled and grouped based on the effect on the vehicle dynamic behaviour.A method is proposed and evaluated, where faults are categorized into different levelsof controllability, i. e. levels on how easy or difficult it is to control a fault for the driver,but also for a control system.Further, fault-tolerant control strategies are suggested that can handle a fault with acritical controllability level. Two strategies are proposed and evaluated based on thecontrol allocation method and an electric vehicle with typical faults. It is shown thatthe control allocation approaches give less critical trajectory deviation compared to noactive control and a regular Electronic Stability Control algorithm.To conclude, this thesis work contributes with a methodology to analyse and developfault-tolerant solutions for electric vehicles with improved traffic safety., QC 20131010

The increase of electronics in road vehicles comes along with a broad variety of possibilitiesin terms of safety, handling and comfort for the users. A rising complexityof the vehicle subsystems and components accompanies this development and has tobe managed by increased electronic control. More potential elements, such as sensors,actuators or software codes, can cause a failure independently or by mutually influencingeach other. There is a need of a structured approach to sort the faults from avehicle dynamics stability perspective.This thesis tries to solve this issue by suggesting a fault classification method and faulttolerantcontrol strategies. Focus is on typical faults of the electric driveline and thecontrol system, however mechanical and hydraulic faults are also considered. Duringthe work, a broad failure mode and effect analysis has been performed and the faultshave been modeled and grouped based on the effect on the vehicle dynamic behaviour.A method is proposed and evaluated, where faults are categorized into different levelsof controllability, i. e. levels on how easy or difficult it is to control a fault for the driver,but also for a control system.Further, fault-tolerant control strategies are suggested that can handle a fault with acritical controllability level. Two strategies are proposed and evaluated based on thecontrol allocation method and an electric vehicle with typical faults. It is shown thatthe control allocation approaches give less critical trajectory deviation compared to noactive control and a regular Electronic Stability Control algorithm.To conclude, this thesis work contributes with a methodology to analyse and developfault-tolerant solutions for electric vehicles with improved traffic safety., QC 20131010

The increase of electronics in road vehicles comes along with a broad variety of possibilitiesin terms of safety, handling and comfort for the users. A rising complexityof the vehicle subsystems and components accompanies this development and has tobe managed by increased electronic control. More potential elements, such as sensors,actuators or software codes, can cause a failure independently or by mutually influencingeach other. There is a need of a structured approach to sort the faults from avehicle dynamics stability perspective.This thesis tries to solve this issue by suggesting a fault classification method and faulttolerantcontrol strategies. Focus is on typical faults of the electric driveline and thecontrol system, however mechanical and hydraulic faults are also considered. Duringthe work, a broad failure mode and effect analysis has been performed and the faultshave been modeled and grouped based on the effect on the vehicle dynamic behaviour.A method is proposed and evaluated, where faults are categorized into different levelsof controllability, i. e. levels on how easy or difficult it is to control a fault for the driver,but also for a control system.Further, fault-tolerant control strategies are suggested that can handle a fault with acritical controllability level. Two strategies are proposed and evaluated based on thecontrol allocation method and an electric vehicle with typical faults. It is shown thatthe control allocation approaches give less critical trajectory deviation compared to noactive control and a regular Electronic Stability Control algorithm.To conclude, this thesis work contributes with a methodology to analyse and developfault-tolerant solutions for electric vehicles with improved traffic safety., QC 20131010

This paper discusses how to use the warping information obtained after batch synchronization for process monitoring and fault classification. The warping information can be used for i) building unsupervised control charts or ii) fault classification when a rich faulty batches database is available. Data from realistic simulations of a fermentation process of the Saccharomyces cerevisiae cultivation are used to illustrate the proposal.

The increase of electronics in road vehicles comes along with a broad variety of possibilitiesin terms of safety, handling and comfort for the users. A rising complexityof the vehicle subsystems and components accompanies this development and has tobe managed by increased electronic control. More potential elements, such as sensors,actuators or software codes, can cause a failure independently or by mutually influencingeach other. There is a need of a structured approach to sort the faults from avehicle dynamics stability perspective.This thesis tries to solve this issue by suggesting a fault classification method and faulttolerantcontrol strategies. Focus is on typical faults of the electric driveline and thecontrol system, however mechanical and hydraulic faults are also considered. Duringthe work, a broad failure mode and effect analysis has been performed and the faultshave been modeled and grouped based on the effect on the vehicle dynamic behaviour.A method is proposed and evaluated, where faults are categorized into different levelsof controllability, i. e. levels on how easy or difficult it is to control a fault for the driver,but also for a control system.Further, fault-tolerant control strategies are suggested that can handle a fault with acritical controllability level. Two strategies are proposed and evaluated based on thecontrol allocation method and an electric vehicle with typical faults. It is shown thatthe control allocation approaches give less critical trajectory deviation compared to noactive control and a regular Electronic Stability Control algorithm.To conclude, this thesis work contributes with a methodology to analyse and developfault-tolerant solutions for electric vehicles with improved traffic safety., QC 20131010

The classification performance of a back propagation neural network classifier highly depends on itstraining process. In this paper we use the domain knowledge stored in a Case-based reasoning system inorder to train a back propagation neural network to classify gear faults in an industrial robot. Ourapproach is to compile domain knowledge from a Case-based reasoning system using attributes frompreviously stored cases. These attributes holds vital information usable in the training process. Ourapproach may be usable when a light-weight classifier is wanted due to e.g. lack of computing power orwhen only a part of the knowledge stored in the case base of a large Case-based reasoning system isneeded. Further, no use of the usual sensor signal classification steps such as filtering and featureextraction are needed once the neural network classifier is successfully trained., Exact

The classification performance of a back propagation neural network classifier highly depends on itstraining process. In this paper we use the domain knowledge stored in a Case-based reasoning system inorder to train a back propagation neural network to classify gear faults in an industrial robot. Ourapproach is to compile domain knowledge from a Case-based reasoning system using attributes frompreviously stored cases. These attributes holds vital information usable in the training process. Ourapproach may be usable when a light-weight classifier is wanted due to e.g. lack of computing power orwhen only a part of the knowledge stored in the case base of a large Case-based reasoning system isneeded. Further, no use of the usual sensor signal classification steps such as filtering and featureextraction are needed once the neural network classifier is successfully trained., Exact

The classification performance of a back propagation neural network classifier highly depends on itstraining process. In this paper we use the domain knowledge stored in a Case-based reasoning system inorder to train a back propagation neural network to classify gear faults in an industrial robot. Ourapproach is to compile domain knowledge from a Case-based reasoning system using attributes frompreviously stored cases. These attributes holds vital information usable in the training process. Ourapproach may be usable when a light-weight classifier is wanted due to e.g. lack of computing power orwhen only a part of the knowledge stored in the case base of a large Case-based reasoning system isneeded. Further, no use of the usual sensor signal classification steps such as filtering and featureextraction are needed once the neural network classifier is successfully trained., Exact

The classification performance of a back propagation neural network classifier highly depends on itstraining process. In this paper we use the domain knowledge stored in a Case-based reasoning system inorder to train a back propagation neural network to classify gear faults in an industrial robot. Ourapproach is to compile domain knowledge from a Case-based reasoning system using attributes frompreviously stored cases. These attributes holds vital information usable in the training process. Ourapproach may be usable when a light-weight classifier is wanted due to e.g. lack of computing power orwhen only a part of the knowledge stored in the case base of a large Case-based reasoning system isneeded. Further, no use of the usual sensor signal classification steps such as filtering and featureextraction are needed once the neural network classifier is successfully trained., Exact

The classification performance of a back propagation neural network classifier highly depends on itstraining process. In this paper we use the domain knowledge stored in a Case-based reasoning system inorder to train a back propagation neural network to classify gear faults in an industrial robot. Ourapproach is to compile domain knowledge from a Case-based reasoning system using attributes frompreviously stored cases. These attributes holds vital information usable in the training process. Ourapproach may be usable when a light-weight classifier is wanted due to e.g. lack of computing power orwhen only a part of the knowledge stored in the case base of a large Case-based reasoning system isneeded. Further, no use of the usual sensor signal classification steps such as filtering and featureextraction are needed once the neural network classifier is successfully trained., Exact

The classification performance of a back propagation neural network classifier highly depends on itstraining process. In this paper we use the domain knowledge stored in a Case-based reasoning system inorder to train a back propagation neural network to classify gear faults in an industrial robot. Ourapproach is to compile domain knowledge from a Case-based reasoning system using attributes frompreviously stored cases. These attributes holds vital information usable in the training process. Ourapproach may be usable when a light-weight classifier is wanted due to e.g. lack of computing power orwhen only a part of the knowledge stored in the case base of a large Case-based reasoning system isneeded. Further, no use of the usual sensor signal classification steps such as filtering and featureextraction are needed once the neural network classifier is successfully trained., Exact

The research presented is motivated by the demand for process improvement for companies active within software development. High demands on software quality are a reality. At the same time, short development time and low effort consumption are required. This stresses the necessity for process improvement. Empirical research methods and close cooperation with the industry partner addressed the research challenge. The research presented in this thesis shows how the analysis of faults through fault classification can be used to determine suitable and required process improvements. Two alternatives are investigated. First, a lightweight approach, and second a fault classification approach targeting all faults. The suitability of the fault classification is stressed as well as the importance of assigning the correct fault class. The latter is determined by classifier agreement calculations. Additionally, the research proposes that the appropriate occasion for a correct fault classification is alleged to be when the fault is corrected. The research also introduces an approach to tailor the verification and validation process. The tailoring process suggested considers the functionality characteristics and the software entity complexity in terms of couplings. This is used to select the appropriate and efficient process for verification and validation.

The research presented is motivated by the demand for process improvement for companies active within software development. High demands on software quality are a reality. At the same time, short development time and low effort consumption are required. This stresses the necessity for process improvement. Empirical research methods and close cooperation with the industry partner addressed the research challenge. The research presented in this thesis shows how the analysis of faults through fault classification can be used to determine suitable and required process improvements. Two alternatives are investigated. First, a lightweight approach, and second a fault classification approach targeting all faults. The suitability of the fault classification is stressed as well as the importance of assigning the correct fault class. The latter is determined by classifier agreement calculations. Additionally, the research proposes that the appropriate occasion for a correct fault classification is alleged to be when the fault is corrected. The research also introduces an approach to tailor the verification and validation process. The tailoring process suggested considers the functionality characteristics and the software entity complexity in terms of couplings. This is used to select the appropriate and efficient process for verification and validation.

The research presented is motivated by the demand for process improvement for companies active within software development. High demands on software quality are a reality. At the same time, short development time and low effort consumption are required. This stresses the necessity for process improvement. Empirical research methods and close cooperation with the industry partner addressed the research challenge. The research presented in this thesis shows how the analysis of faults through fault classification can be used to determine suitable and required process improvements. Two alternatives are investigated. First, a lightweight approach, and second a fault classification approach targeting all faults. The suitability of the fault classification is stressed as well as the importance of assigning the correct fault class. The latter is determined by classifier agreement calculations. Additionally, the research proposes that the appropriate occasion for a correct fault classification is alleged to be when the fault is corrected. The research also introduces an approach to tailor the verification and validation process. The tailoring process suggested considers the functionality characteristics and the software entity complexity in terms of couplings. This is used to select the appropriate and efficient process for verification and validation.

The research presented is motivated by the demand for process improvement for companies active within software development. High demands on software quality are a reality. At the same time, short development time and low effort consumption are required. This stresses the necessity for process improvement. Empirical research methods and close cooperation with the industry partner addressed the research challenge. The research presented in this thesis shows how the analysis of faults through fault classification can be used to determine suitable and required process improvements. Two alternatives are investigated. First, a lightweight approach, and second a fault classification approach targeting all faults. The suitability of the fault classification is stressed as well as the importance of assigning the correct fault class. The latter is determined by classifier agreement calculations. Additionally, the research proposes that the appropriate occasion for a correct fault classification is alleged to be when the fault is corrected. The research also introduces an approach to tailor the verification and validation process. The tailoring process suggested considers the functionality characteristics and the software entity complexity in terms of couplings. This is used to select the appropriate and efficient process for verification and validation.

The research presented is motivated by the demand for process improvement for companies active within software development. High demands on software quality are a reality. At the same time, short development time and low effort consumption are required. This stresses the necessity for process improvement. Empirical research methods and close cooperation with the industry partner addressed the research challenge. The research presented in this thesis shows how the analysis of faults through fault classification can be used to determine suitable and required process improvements. Two alternatives are investigated. First, a lightweight approach, and second a fault classification approach targeting all faults. The suitability of the fault classification is stressed as well as the importance of assigning the correct fault class. The latter is determined by classifier agreement calculations. Additionally, the research proposes that the appropriate occasion for a correct fault classification is alleged to be when the fault is corrected. The research also introduces an approach to tailor the verification and validation process. The tailoring process suggested considers the functionality characteristics and the software entity complexity in terms of couplings. This is used to select the appropriate and efficient process for verification and validation.

The research presented is motivated by the demand for process improvement for companies active within software development. High demands on software quality are a reality. At the same time, short development time and low effort consumption are required. This stresses the necessity for process improvement. Empirical research methods and close cooperation with the industry partner addressed the research challenge. The research presented in this thesis shows how the analysis of faults through fault classification can be used to determine suitable and required process improvements. Two alternatives are investigated. First, a lightweight approach, and second a fault classification approach targeting all faults. The suitability of the fault classification is stressed as well as the importance of assigning the correct fault class. The latter is determined by classifier agreement calculations. Additionally, the research proposes that the appropriate occasion for a correct fault classification is alleged to be when the fault is corrected. The research also introduces an approach to tailor the verification and validation process. The tailoring process suggested considers the functionality characteristics and the software entity complexity in terms of couplings. This is used to select the appropriate and efficient process for verification and validation.